Following the introduction of highly active antiretroviral therapy (HAART) and improved survival among HIV-infected persons, non-infectious lung diseases are hypothesized to increase as causes of morbidity and mortality. Limited evidence suggests that HIV-infected persons may be more susceptible to development of emphysema. An increase in lung cancer risk has been observed with HIV infection;we have demonstrated that this risk persists after controlling for heavy tobacco use, among HIV-infected persons. HIV infection is recognized as an independent risk factor for pulmonary hypertension;however accurate characterization of the prevalence and clinical manifestations remains elusive. To address these issues, we propose SHIELD: The Study of HIV Infection in the Etiology of Lung Disease, a collaborative and systematic investigation of lung disease among HIV-infected persons nested within three on-going HIV cohort studies in Baltimore, MD. Our specific aims include: 1) to determine the prevalence, clinical manifestations, and risk factors for chronic obstructive pulmonary disease (COPD) among HIV-infected persons. Prospective evaluation will characterize the effect of HIV infection, immune deficiency, HAART use, pulmonary infections, smoking, and illicit drug use on rates of decline in lung function and on related morbidity and mortality;2) To determine incidence rates, clinical manifestations, risk factors, and mechanisms for lung cancer among HIV-infected persons, and 3) To determine the prevalence, clinical manifestations, and risk factors for pulmonary arterial hypertension (PAH) among HIV-infected persons. The collective experience of our HIV cohorts spans well over two decades and includes follow-up on more than 11,000 participants;active follow-up of over 4,000 HIV-infected persons are available for inclusion in focused studies of COPD, lung cancer, and PAH. SHIELD represents a consortium of investigators with multidisciplinary expertise in pulmonary medicine, HIV and infectious diseases, surgical oncology, epidemiology and biostatistics, translational imaging techniques, immunology and basic sciences. Given the established collaborative research team, the substantive existing resources and infrastructure, and the strength of our scientific methods, we anticipate a high likelihood of success in characterizing non-infectious lung disease in HIV infection and contributing to an improved understanding of mechanisms involved.